The manual reads: Mvasc3= 3-phase short circuit MVA= kVBase^2/Zsc
Question: In which unit must Mvasc3 and Mvasc1 be input? kVA or MVA?
Question: In Germany, short circuit power (3phase) is sqrt(3)*Un*Ik , where Un is the nominal voltage (phase to phase) and Ik is the short circuit current in EACH of the phases, and Zk is defined as (Un/sqrt(3))/Ik
What does Zsc stand for in OpenDSS?
or, in short, if I have a voltage source with short circuit power sqrt(3)*Un*Ik=50MVA,
which Mvasc3 and Mvasc1 values must I specify in OpenDSS?
My guess is Mvasc3=50 and Mvasc1=50/sqrt(3)=28.87 -- is this correct?
(Edited 2016-05-19 7:55UTC: escape the asterisks which are intended as multiplication signs, but in the original post resulted in markup for italic.)
Last edit: Wilfried Hennings 2016-05-19
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
MVAsc3 and MVasc1 must be defined in MVA. For example, MVAsc1 = 2100 means that the single-phase short circuit power is 2100MVA. Note that this value stands for the module of this parameter (|MVAsc3|).
Zsc stands for the positive sequence impedance (also called Z1) of the voltage source. You can check it in the help menu: "MVA Short circuit, 3-phase fault. Default = 2000. Z1 is determined by squaring the base kv and dividing by this value".
In your case, you must specify MVAsc3 = 50. However, you can't find MVAsc1 directly from MVAsc3. They're independent from each other and you need both of them to describe the Vsource.
In OpenDSS you have three options to define your Vsource element:
1. By using Z1=[R1,X1] and Z0=[R0,X0]
2. Through MVAsc3 and MVasc1, which means |MVAsc3| and |MVAsc1|.
3. Through Isc3 and Isc1, which means |Isc3| and |Isc1|.
For the 2nd and the 3rd case, OpenDSS uses the parameters x1r1 and x0r0 to define X1/R1 and X0/R0 ratios, respectively. As OpenDSS performs the powerflow using three-phase parameters, in its internal algorithm, it converts (Z1,Z0), (MVAsc3,MVAsc1) and (Isc3,Isc1) to a series equivalent impendance, Zs, and a mutual impedance, Zm.
Hope it helps!
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
OK, thank you! I decided to specify R1 X1 R0 X0, which are (hopefully) unambiguous.
A scholar's textbook sais that for a Dyn transformer the null resistance and inductance are equal to the positive sequence resistance and inductance, resp.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I assume by "null" you mean zero-sequence. Maybe the terminology is a little different.
Zsc is the short-circuit (sc) impedance and we generally use two of them to define short-circuit (Thevinen) equivalents for distribution system analysis: the 3-phase-to-earth impedance and 1-phase-to-earth impedance. It is really a 3x3 matrix, but that is difficult to get because transmission system analysis software seldom uses full 3-phase modeling. So we settle for the symmetrical component, or sequence impedances, because those are typically available in some form -- either as the short-circuit MVA for 3-phase and 1-phase fault or the short-circuit current magnitudes for 3-phase and 1-phase faults.
The textbook you refer to is partially correct with respect to the impedances of a Dyn transformer. It all depends on the core configuration of the transformer. However, the bigger issue may be where you are going to make your equivalent. If you make the equivalent at the MV bus in a typical substation, the positive-sequence impedance is often higher than the zero-sequence impedance looking back into the MV terminals of the transformer. Of course, there is an exception if there is a large neutral reactor (or Petersen coil) connected to the yn winding. If you include that reactor in the equivalent, the zero-sequence impedance will be much higher. If you intend to model the neutral reactor separately connected to the neutral point of the Vsource (terminal 2) then you will use the lower value. So it all depends.
I generally use the Vsource impedance to model only the bulk power system (HV) impedance up to the HV terminals of the transformer. Then use the transformer's impedance for the Transformer object as a separate element. If I have to be more exact (not often), I will add a grounding transformer (wye-delta or zig-zag) in the model at the LV terminals to lower the effective Z0 looking back into the transformer. The 3-limb core design of the substation transformer will make it appear as if the transformer has a 3rd delta-connected winding with a high impedance. This is sometimes called the "phantom tertiary", which you can find in various transformer text books. There is also some information on the OpenDSS Doc folder on this subject if you want to know more. If you used the OpenDSSInstraller to install the program, you have it on your computer.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
This is indeed one source of problems, but different definitions of parameters make more problems.
You are right that it would be better to model vsource on the HV level (110kV) and model the HV-MV transformer explicitly (MV is 10kV). However I first want to recalculate a case given in a publication where vsource is just feeding the MV-LV substation transformer (LV is 0.4kV). The publication lists only the three-phase short circuit power 50MVA and R/X=0.7 at the MV side of the MV-LV transformer and neither lists the single phase short circuit power nor the zero-sequence impedance.
I hope that Edit Vsource.Source BasekV=10 pu=1.0 R1=1.1469 X1=1.6385 R0=1.1469 X0=1.6385
will model the given data correctly, as
Z = Un²/Sk = 10000²/50000000 = 2
sqrt(1.1469² + 1.6385²) = 2
1.1469 / 1.6385 = 0.7
Last edit: Wilfried Hennings 2016-05-19
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
It is probably a high-impedance grounded MV system. Since the MV/LV transformer is Dyn, it probably doesn't matter what you put for the 1-phase short circuit. I would probably put in a small value of 1-5 A for Isc1 and that will likely suffice for simulations.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
The manual reads: Mvasc3= 3-phase short circuit MVA= kVBase^2/Zsc
Question: In which unit must Mvasc3 and Mvasc1 be input? kVA or MVA?
Question: In Germany, short circuit power (3phase) is sqrt(3)*Un*Ik , where Un is the nominal voltage (phase to phase) and Ik is the short circuit current in EACH of the phases, and Zk is defined as (Un/sqrt(3))/Ik
What does Zsc stand for in OpenDSS?
or, in short, if I have a voltage source with short circuit power sqrt(3)*Un*Ik=50MVA,
which Mvasc3 and Mvasc1 values must I specify in OpenDSS?
My guess is Mvasc3=50 and Mvasc1=50/sqrt(3)=28.87 -- is this correct?
(Edited 2016-05-19 7:55UTC: escape the asterisks which are intended as multiplication signs, but in the original post resulted in markup for italic.)
Last edit: Wilfried Hennings 2016-05-19
MVAsc3 and MVasc1 must be defined in MVA. For example, MVAsc1 = 2100 means that the single-phase short circuit power is 2100MVA. Note that this value stands for the module of this parameter (|MVAsc3|).
Zsc stands for the positive sequence impedance (also called Z1) of the voltage source. You can check it in the help menu: "MVA Short circuit, 3-phase fault. Default = 2000. Z1 is determined by squaring the base kv and dividing by this value".
In your case, you must specify MVAsc3 = 50. However, you can't find MVAsc1 directly from MVAsc3. They're independent from each other and you need both of them to describe the Vsource.
In OpenDSS you have three options to define your Vsource element:
1. By using Z1=[R1,X1] and Z0=[R0,X0]
2. Through MVAsc3 and MVasc1, which means |MVAsc3| and |MVAsc1|.
3. Through Isc3 and Isc1, which means |Isc3| and |Isc1|.
For the 2nd and the 3rd case, OpenDSS uses the parameters x1r1 and x0r0 to define X1/R1 and X0/R0 ratios, respectively. As OpenDSS performs the powerflow using three-phase parameters, in its internal algorithm, it converts (Z1,Z0), (MVAsc3,MVAsc1) and (Isc3,Isc1) to a series equivalent impendance, Zs, and a mutual impedance, Zm.
Hope it helps!
OK, thank you! I decided to specify R1 X1 R0 X0, which are (hopefully) unambiguous.
A scholar's textbook sais that for a Dyn transformer the null resistance and inductance are equal to the positive sequence resistance and inductance, resp.
I assume by "null" you mean zero-sequence. Maybe the terminology is a little different.
Zsc is the short-circuit (sc) impedance and we generally use two of them to define short-circuit (Thevinen) equivalents for distribution system analysis: the 3-phase-to-earth impedance and 1-phase-to-earth impedance. It is really a 3x3 matrix, but that is difficult to get because transmission system analysis software seldom uses full 3-phase modeling. So we settle for the symmetrical component, or sequence impedances, because those are typically available in some form -- either as the short-circuit MVA for 3-phase and 1-phase fault or the short-circuit current magnitudes for 3-phase and 1-phase faults.
The textbook you refer to is partially correct with respect to the impedances of a Dyn transformer. It all depends on the core configuration of the transformer. However, the bigger issue may be where you are going to make your equivalent. If you make the equivalent at the MV bus in a typical substation, the positive-sequence impedance is often higher than the zero-sequence impedance looking back into the MV terminals of the transformer. Of course, there is an exception if there is a large neutral reactor (or Petersen coil) connected to the yn winding. If you include that reactor in the equivalent, the zero-sequence impedance will be much higher. If you intend to model the neutral reactor separately connected to the neutral point of the Vsource (terminal 2) then you will use the lower value. So it all depends.
I generally use the Vsource impedance to model only the bulk power system (HV) impedance up to the HV terminals of the transformer. Then use the transformer's impedance for the Transformer object as a separate element. If I have to be more exact (not often), I will add a grounding transformer (wye-delta or zig-zag) in the model at the LV terminals to lower the effective Z0 looking back into the transformer. The 3-limb core design of the substation transformer will make it appear as if the transformer has a 3rd delta-connected winding with a high impedance. This is sometimes called the "phantom tertiary", which you can find in various transformer text books. There is also some information on the OpenDSS Doc folder on this subject if you want to know more. If you used the OpenDSSInstraller to install the program, you have it on your computer.
"Maybe the terminology is a little different."
This is indeed one source of problems, but different definitions of parameters make more problems.
You are right that it would be better to model vsource on the HV level (110kV) and model the HV-MV transformer explicitly (MV is 10kV). However I first want to recalculate a case given in a publication where vsource is just feeding the MV-LV substation transformer (LV is 0.4kV). The publication lists only the three-phase short circuit power 50MVA and R/X=0.7 at the MV side of the MV-LV transformer and neither lists the single phase short circuit power nor the zero-sequence impedance.
I hope that
Edit Vsource.Source BasekV=10 pu=1.0 R1=1.1469 X1=1.6385 R0=1.1469 X0=1.6385will model the given data correctly, as
Z = Un²/Sk = 10000²/50000000 = 2
sqrt(1.1469² + 1.6385²) = 2
1.1469 / 1.6385 = 0.7
Last edit: Wilfried Hennings 2016-05-19
It is probably a high-impedance grounded MV system. Since the MV/LV transformer is Dyn, it probably doesn't matter what you put for the 1-phase short circuit. I would probably put in a small value of 1-5 A for Isc1 and that will likely suffice for simulations.